一个Tensorflow搭建的网络可视化的例子(有代码)

[复制链接]
666|3
手机看帖
扫描二维码
随时随地手机跟帖
keer_zu|  楼主 | 2019-1-10 18:47 | 显示全部楼层 |阅读模式


将代码运行,会在本地产生一个:logs\ 的文件夹,执行:
tensorboard --logdir  logs


然后出现:

TensorBoard 1.12.0 at http://ubuntu:6006 (Press CTRL+C to quit)


在本机浏览器上输入:

http://ubuntu:6006
即可看到可视化的效果。


代码:
from __future__ import print_function 
import tensorflow as tf
import numpy as np

def add_layer(inputs, in_size, out_size, n_layer, activation_function=None):
# add one more layer and return the output of this layer
    layer_name = 'layer%s' % n_layer
    with tf.name_scope(layer_name):
        with tf.name_scope('weights'):
            Weights = tf.Variable(tf.random_normal([in_size, out_size]), name='W')
            tf.summary.histogram(layer_name + '/weights', Weights)
        with tf.name_scope('biases'):
            biases = tf.Variable(tf.zeros([1, out_size]) + 0.1, name='b')
            tf.summary.histogram(layer_name + '/biases', biases)
        with tf.name_scope('Wx_plus_b'):
            Wx_plus_b = tf.add(tf.matmul(inputs, Weights), biases)
        if activation_function is None:
            outputs = Wx_plus_b
        else:
            outputs = activation_function(Wx_plus_b, )
        tf.summary.histogram(layer_name + '/outputs', outputs)
    return outputs





# Make up some real data
x_data = np.linspace(-1, 1, 300)[:, np.newaxis]
noise = np.random.normal(0, 0.05, x_data.shape)
y_data = np.square(x_data) - 0.5 + noise



# define placeholder for inputs to network
with tf.name_scope('inputs'):
    xs = tf.placeholder(tf.float32, [None, 1], name='x_input')
    ys = tf.placeholder(tf.float32, [None, 1], name='y_input')




# add hidden layer
l1 = add_layer(xs, 1, 10, n_layer=1, activation_function=tf.nn.relu)



# add output layer
prediction = add_layer(l1, 10, 1, n_layer=2, activation_function=None)


# the error between prediciton and real data
with tf.name_scope('loss'):
    loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction), reduction_indices=[1]))
    tf.summary.scalar('loss', loss)


with tf.name_scope('train'):
    train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

sess = tf.Session()
merged = tf.summary.merge_all()
writer = tf.summary.FileWriter("logs/", sess.graph)
init = tf.global_variables_initializer()
sess.run(init)

for i in range(1000):
    sess.run(train_step, feed_dict={xs: x_data, ys: y_data})
    if i % 50 == 0:
        result = sess.run(merged, feed_dict={xs: x_data, ys: y_data})
        writer.add_summary(result, i)



相关帖子

keer_zu|  楼主 | 2019-1-11 08:48 | 显示全部楼层
@gaoyang9992006 可以玩玩tensorflow

使用特权

评论回复
发新帖 我要提问
您需要登录后才可以回帖 登录 | 注册

本版积分规则

个人签名:qq群:49734243 Email:zukeqiang@gmail.com

1288

主题

12164

帖子

53

粉丝